Study of Online Bayesian Networks Learning in a Multi-Agent System

Provided by: International Journal of Computer Science Issues
Topic: Mobility
Format: PDF
The authors attempt to find how a common task can be performed by a multi-agent self-organizing system. The agents are independent in terms of their model of environment and their actions. Each agent explores the environment and decides its actions by itself. Agents will have no information about the environment at the beginning of their exploration of the environment. This paper introduces online Bayesian network learning in detail. The structural and parametric learning abilities of the online Bayesian network learning are explored. The paper starts with revisiting the multi-agent self-organization problem and the proposed solution. Then, they explain the proposed Bayesian network learning, three scoring functions, namely Log-Likelihood, Minimum description length, and Bayesian scores.

Find By Topic